Age-related macular degeneration (AMD) is the leading cause of blindness in older people. Inflammation, mediated by activated microglia/macrophages (mi/ma), plays a key role in the pathogenesis of both geographic atrophy and exudative AMD, in humans and in animal models. Development of systemic drug delivery approaches targeting these cells to treat ocular inflammation and choroidal neovascularization (CNV), can have significant benefits, in addition to providing insights on the role of mi/ma on AMD therapy. Such therapies may improve patient compliance, reduce side effects arising from intravitreal implants and free drugs, reduce costs, and allow therapies during early stages of AMD. Understanding and utilizing the transport of nanoparticles from blood to the choroid/retina may provide new avenues for systemic drug delivery. We explored globular, polyamidoamine (PAMAM) dendrimers, with favorable physicochemical attributes. Our preliminary results in a lipid-induced rat dry/wet AMD model demonstrate that systemic hydroxyl-functionalized PAMAM dendrimers: (1) can target activated mi/ma in the areas of choroidal neovascularization and retinal inflammation, and be selectively retained for a sustained period of time; (2) can deliver anti-inflammatory, anti- oxidant N-acetyl cysteine (NAC) to attenuate proinflammatory cytokines, and cause significant CNV suppression, when administered in early stages (day 3 after lipid, `dry' AMD); (3) deliver a combination of NAC and triamcinolone acetonide (TA), to cause CNV regression, when administered in late stages (day 11 after lipid, `wet' AMD); We explore three aims: (1) Determine whether systemic dendrimer-drug conjugates selectively localize and retain in activated microglia/macrophages in the choroid/retina, and whether an increase in dendrimer size improves such localization and retention in the rat AMD model. (2) Determine the ocular toxicity of the systemic dendrimer vehicle in healthy rats, and characterize pharmacokinetics and biodistribution of the conjugated drugs in the rat CNV model. (3) Determine if systemically administered D-NAC and/or D-TA conjugates attenuate inflammation, suppress and cause regression of CNV in the rat AMD model. If successful, these studies offer potential for targeted therapies for dry/wet AMD, and broad clinical adaptation.